Jan vs Locally AI

Jan

• Running open models offline on your own computer
• People who want one desktop app for local and cloud models
• Developers who want a local OpenAI-compatible endpoint

Locally AI

• Trying open models on an iPhone, iPad or Mac at no cost
• LM Studio users who want their desktop models on a phone
• Private chats that work with no connection and no account

Jan

• Desktop only, on macOS, Windows and Linux. Intel-based Macs are not supported.
• Local models need memory: Jan's Mac guide suggests 8GB of RAM for 3B models, 16GB for 7B models and 32GB for 13B models.
• Cloud models need your own API key from each provider, and those providers see the messages you send them.
• Cowork and the standalone Jan Agent are previews on nightly builds, and the memory feature is marked as coming soon.

Locally AI

• It is available only for recent iPhone, iPad and Mac models, and is optimized for Apple Silicon chips.
• Which models run depends on the device's memory; the site points to iPad and Mac for the largest ones.
• Using larger models through LM Link requires LM Studio running on a computer of your own.
• LM Studio acquired the app in April 2026, and the App Store now lists it as Locally AI by LM Studio.

Jan

Local models

Downloads and runs open models such as Llama, Gemma and Qwen from Hugging Face using llama.cpp or MLX. The Hub shows whether each model fits your hardware.

Cloud providers with your own key

Connects to OpenAI, Anthropic, Gemini, Mistral AI, Groq, OpenRouter and Azure OpenAI, or to any OpenAI- or Anthropic-compatible endpoint, using an API key you supply.

Built-in web search

Web search and page fetch tools are on by default for models that support tool use. They run through a keyless Exa endpoint, so no API key or MCP server is needed.

Projects and assistants

Projects group conversations under shared instructions and files, and custom assistants keep their own instructions for recurring kinds of work.

Local API server

A built-in OpenAI-compatible server at localhost:1337 lets other applications use the models running in Jan, and Claude Code can be set to use them as its backend.

MCP connectors and Cowork preview

MCP servers add tools such as code execution and databases. Cowork, which works through multi-step tasks on files and commands, ships only in nightly builds as a preview.

Locally AI

On-device models through MLX

Downloads and runs open models from the Llama, Gemma, Qwen, DeepSeek, SmolLM, Granite, Cogito and LFM families on Apple Silicon, using Apple's MLX framework.

Offline, with no account

Once a model is downloaded, chats run with no internet connection and no login. The vendor says processing stays on the device and no data is collected.

LM Link to LM Studio

Links the app to LM Studio on your own computer over an end-to-end encrypted connection, so an iPhone or iPad can use larger models than it could run itself.

Local voice mode

Holds real-time spoken conversations with a model, with the speech handled on the device instead of by a cloud service.

Language and vision models

Some models take images as well as text, so a chat can analyze a picture. The app also supports Apple's on-device Foundation Models on current OS releases.

Siri, Shortcuts and Control Center

Talk to a local model through Siri, call models from the Shortcuts app to build automations, and open the app from Control Center, the Lock Screen or the Action Button.

Jan

• Free$0

  The full desktop app for macOS, Windows and Linux, with local models, cloud provider connections and the local API server.

Locally AI

• Free$0

  The full app for iPhone, iPad and Mac: local models, voice mode, Siri and Shortcuts actions, and LM Link to LM Studio.